Modeling of Nonlinear Load Electric Energy Measurement and Evaluation System Based on Artificial Intelligence Algorithm

Abstract

Background: To improve the modeling efficiency of nonlinear load electric energy metering evaluation system, a method based on artificial intelligence algorithm was proposed. Methods: First introduces the artificial glowworm swarm optimization extreme learning machine, a powerful ability of global optimization using artificial firefly algorithm. Then, find the hours of training error, extreme learning machine model, input weighting matrix, and hidden layer offset matrix. Moreover, there is a certain period in a given area in our country, power load simulation through the experiment, verifying the validity and superiority of the model. Results: The experimental results show that the traditional BP neural network has the largest prediction relative error, and the stability of BP neural network is poor, and the relative error time is large, which is related to the defect of the neural network itself. The prediction effect of SVM method is better than that of BP neural network, because SVM has a strict theoretical and mathematical basis, so its generalization ability is better than that of BP neural network, and the algorithm has global optimality. Conclusion: As can be seen from the chart analysis, GSO-ELM algorithm performs better than both in terms of stability and test error. It is proved that the modeling of nonlinear load electrical energy measurement and evaluation system based on artificial intelligence algorithm is superior and effective. The proposed algorithm outperforms very well over the existing literature.